In the beginning of this week, Microsoft announced its preparation towards continuing discussions for purchase of TikTok in the United States, following ban threats in the country. This move is in response to concerns shown by President Trump regarding its privacy practices and possible connections to Chinese government.
Recently, TikTok has decided to establish new moderation policies for better protection of its platform against wrong information, electoral interference and other manipulative content ahead of this year’s election. The company now says that it bans all ‘deepfakes’ explicitly. Deepfakes are AI-powered manipulations of video and audio designed for misleading people about something a person may have said or done.
In a blog post published by Vanessa Pappas on Wednesday, it was mentioned that the company is adding a policy which will prohibit manipulated or synthetic content which misleads users by distortion of truth ‘in a way that could cause harm’. She further elaborated that their intent is of protecting users from deep or shallow fakes.
Deepfakes in TikTok have commonly been associated with face-swapping videos which are used for creating pornographic content. Until now they have not been used by a major political campaign, although President Trump’s re-election campaign and White House have shared or created lesser sophisticated edits and other similar deceptive content that has been labeled very misleading.
In recent years, we have observed an increase in techniques of sophistication and ease of usage in creating these deepfakes. This technology has greatly raised concerns regarding how deepfakes may at some point be used to make deceptive edits of politicians expressing support or saying things which may harm their reputation or discredit them. Several major social media platforms and some states have completely banned deepfakes in political advertisements consequently.
TikTok does not allow any political ads already. The company adds that its deepfake ban is intended towards making it harder to use the platform towards pushing deceptive media for political gains. According to its new moderation policies, Tiktok is banning use of bot and fake accounts which mislead people regarding identity of the account holder with an intention of swaying public opinion or exerting other forms of influence. It is also expanding its fact-checking partnerships with Lead Stories and PolitiFact for fact-checking of potential misinformation related to US election 2020. It is also adding an election misinformation option to its in-app reporting mechanism for letting users flag suspicious accounts or content. The app will also contain a new ‘election information center’ for pointing users towards authentic information on voting, race and other related topics.